Yaron Naymark

In investing, the best way to get to yes is by running out of reasons to say no

What part of the cycle are we in when lumberyards trade for 10.7x $QXO is acquiring Kodiak Building Partners for $2.25B Kodiak generated ~$2.4B revenue / ~$211M EBITDA Few years ago I was on the sell side for a lumberyard and we were lucky to get 6x PE-backed consolidating platforms targeted 8x upon exit so Kodiak trading for 10x+ is extraordinary

The below is according to grok. Why/where is this wrong? Once the GSEs hit their regulatory capital target (ERCF or any lowered version), the 2021 PSPA amendment kicks in: • The GSEs pay quarterly dividends to Treasury equal to the lesser of: • 10% of the senior preferred liquidation preference (annual rate), or • The quarterly increase in the GSE’s net worth (i.e., basically that quarter’s net income). • 2025 actual combined net income: $25.1 billion ($14.4B Fannie + $10.7B Freddie). • Current senior preferred liquidation preference: ≈ $373 billion combined at year-end 2025 (it grew by the full $25.1B retained earnings in 2025; it was ~$348B earlier in the year). • 10% of $373B = $37.3 billion per year — larger than the $25.1B in earnings. Result: The binding limit is the net-worth-increase cap, so Treasury receives essentially 100% of ongoing net income as dividends each quarter. Government makes ≈ $25 billion per year (recurring cash dividends). The GSEs’ net worth stays flat at the capital target; all future profits flow to Treasury. No 25 bps fee exists, and there are no common dividends or warrant monetization.

The math on Fannie and Freddie is so dislocated it looks like a pricing error. Fannie printed $14.4 billion in net income last year. Freddie printed $10.7 billion. Combined market cap on the pink sheets right now: ~$12 billion. The market is pricing $25 billion in annual earnings at a 0.48x multiple. Find me another 0.48x earnings multiple anywhere in American finance. It doesn't exist. The dilution fear is the reason the stock is cheap and the reason the stock is wrong. Treasury put in $187 billion. The GSEs have swept back over $300 billion since 2012. That's an 11.6% IRR. If Treasury exercises its 79.9% warrants at today's price, the government's stake is worth ~$9.6 billion. If it exercises post-relist at 10x earnings, that stake is worth $200 billion. The difference is $190 billion. Washington doesn't leave $190 billion on the table to spite penny stock holders. Capital requirements look scary until you do the arithmetic. The ERCF says $334 billion. They have $179 billion. The FHFA can lower Tier 1 to 2.5% without Congress. New target: ~$190 billion. Gap: $11 billion. One IPO closes it. One year of retained earnings closes it twice. G-fees are already at 65 bps. Pre-crisis they were 20. The GSEs have been charging privatized pricing inside a conservatorship for 14 years. Credit losses outside of 2008 average under 5 bps. The margin is so fat that mortgage rates don't move at all on release. So what are you actually buying at $5? A royalty on the American mortgage system. 65 bps on $7.5 trillion in outstanding MBS. $48 billion in gross annual revenue. Under 5 bps in historical losses. The most predictable spread in finance, backstopped by a guarantee both parties have publicly committed to preserving. JPMorgan trades at 13x and takes real credit risk. Utilities trade at 15x with half the visibility. These two trade at 0.48x collecting tolls on other people's risk. The second those warrants convert and the NYSE listing goes live, every index fund and pension fund with a financial sector mandate has to buy. Two of the ten most profitable companies in America, sitting on the pink sheets, waiting for one signature.